Digital & IT
Software in biology is honest exactly as far as what fed its model is visible; usually it was fed less than the demo promises.
The cluster survived triage by the sector’s rule: workflow shells and registries stayed out; the pages that remain are where software meets biology and loses in predictable places. A fermentation model is data-starved by construction — a bioprocess does not repeat enough to learn from. The design–build–test–learn loop is limited by its slowest, noisiest step, and that is not an algorithm. A genome cannot be anonymised, and it is not only yours. Synthetic-DNA screening is a similarity search with a defeatable threshold. A biologics plant cannot be sped up: a batch has a biological duration.
Start with AI strain-design platforms: the loop-is-limited-by-measurement logic ties the cluster together.
Articles 10
| Agricultural bioinformatics cloud platforms Why a breeder can predict a plant's value from its DNA before growing it — and why the prediction expires. |
| AI for clinical trial patient recruitment Why matching a patient to a protocol is a reasoning problem, not a search problem — and why the record it reasons over was written for billing. |
| Biosecurity in cyberspace Why screening a DNA order is a similarity search with a defeatable threshold — and why lab instruments are the hardest computers on a network to patch. |
| Biopharmaceutical production scheduling Why a biologics plant cannot be sped up: a batch has a biological duration, and every changeover has a validated one. |
| Genomic data privacy platforms A genome cannot be anonymised, and it is not only yours — which is why privacy here is cryptography rather than redaction. |
| AI bioprocess optimisation as a service Why a model of a fermentation is data-starved by construction — and what changes when the model lives off-site. |
| AI strain design and engineering platforms The design–build–test–learn loop is limited by its slowest, noisiest step — and that step is not the design. |
| Bio-inspired AI architectures The energy argument for spiking silicon is real and narrow — and the training problem behind it has no biological answer. |
| Calibration management systems Scheduling is the visible half; the substance is an unbroken chain of comparisons and a statistical bet on drift. |
| Quantum computing for molecular modelling The problem is real, the mapping is elegant, and no existing device has beaten a classical method on any molecule anyone cares about. |